Campbell Scientific CR510 Basic Datalogger User Manual
Page 180

APPENDIX B. BINARY TELECOMMUNICATIONS
B-4
LO RESOLUTION FORMAT - D,E,F,
NOT ALL ONES
Bits
Description
A
Polarity, 0 = +, 1 = -.
B, C
Decimal locators as defined below.
D-H plus
13 bit binary value (D=MSB).
second
Largest possible number without D,
E, and F all 1 is 7167,
byte
but CSI defines the largest
allowable range as 6999.
The decimal locators can be viewed as a
negative base 10 exponent with decimal
locations as follows:
B
C
Decimal Location
0
0
XXXX.
0
1
XXX.X
1
0
XX.XX
1
1
X.XXX
DATA TYPE WHEN D,E,F, ALL EQUAL ONE
If D, E, and F are all ones, the data type is
determined by the other bits as shown below. X
implies a "don't care" condition; i.e., the bit can
be either 1 or 0 and is not used in the decode
decision.
A
B
C
D
E
F
G
H
DATA TYPE AND SECOND BYTE FORMAT
1
1
1
1
1
1
X
X
A,B,C, = 1 - Start of Output Array, G & H are the most
significant bits of the Output Array ID. All 8 bits of the
2nd byte are also included in the ID.
X
X
0
1
1
1
X
X
C = 0 - First byte of a 4 byte value.
0
0
1
1
1
1
X
X
A,B = 0; C = 1 - Third byte of a 4 byte value.
0
1
1
1
1
1
1
1
A = 0; remaining bits = 1 - First byte of a 2 byte
"dummy" word. The CR510 always transmits a 0 for
the 2nd byte, but the word can be decoded on the basis
of the 1st byte only.
HI RESOLUTION FORMAT
Continuing to use the A-H bit representation, the four byte number is shown below as two two byte pairs.
AB0111GH
XXXXXXXX
001111GH
XXXXXXXX
BITS, 1ST BYTE,
1ST PAIR
DESCRIPTION
CDEF = 0111
Code designating 1st byte pair of four byte number.
B
Polarity , 0 = +, 1 = -.
G,H,A,
Decimal locator as defined below.
2nd byte
16th - 9th bit (left to right) of 17 bit binary value.
ABCDEF = 001111
Code designating 2nd byte pair of four byte number.
G
Unused bit.
H
17th and MSB of 17 bit binary value.
2nd byte
8th - 1st bit (left to right) of 17 bit binary value.